Introduction

Inherited neuropathies are common disorders with a prevalence of at least 1 in 2500 individuals.1 The classification is based on clinical features and electrophysiology, but over the last few years, genetic testing has shown that even clinically identical inherited neuropathies can be caused by different genetic defects.2, 3, 4, 5, 6, 7, 8 Pupil abnormalities are an important aspect of the clinical characterization of patients with inherited neuropathy, but they have been investigated only in isolated cases or small groups.9 Please refer to the Supplementary text, which has an extensive review of the literature on pupil abnormalities, as only brief background information is given here.

Familial amyloid polyneuropathy (FAP) is an inherited neuropathy with autonomic and systemic manifestations.10 In FAP, there is deposition of amyloid material within the autonomic nerve supply of the ocular tissues, predominantly the vitreous body, conjunctivae, and ciliary ganglia,11 with the development of scalloped pupils.12, 13 The pupil abnormalities in FAP have also been identified in immunoglobulin light-chain-associated amyloidosis,14 but only a small number of the commonest Transthyretin FAP mutation types have been investigated.14, 15, 16 Abnormal ocular signs are also an important feature of Refsum's disease as in Refsum's original report17 where the pupils were miotic and responded poorly to light, better to near, and dilated poorly in response to homatropine eyedrops. Similar observations have been made by subsequent authors,18 but only in single cases or small series. The neurological phenotype of hereditary sensory and autonomic neuropathy (HSAN) is similar to FAP, but there have been no cases or families reported examined in detail.19

In Charcot Marie Tooth disease (CMT), pupil abnormalities have been reported over the years but mainly in single families,9 and most cases were diagnosed with the clinically severe form of CMT called Déjérine–Sottas syndrome. Recently, pupil abnormalities have been shown to be particularly prevalent in certain myelin protein zero (MPZ) gene mutations,20, 21 and sluggish pupillary responses to light and near have been reported in a Gypsy family with HMSN-Lom.22 No reports have analysed the pupils in hereditary neuropathy with liability to pressure palsies (HNPP).

In this study, we have analysed the genetic aetiology and the pupil abnormalities in a prospective series of 131 cases and cases presenting between 1998 and 2007 with inherited neuropathy. The patients seen formed five broad groups of inherited neuropathy: FAP, CMT, HNPP, HSAN, and Refsum's disease.

Materials and methods

Patients

Ethics approval was obtained from the joint medical and ethics committee at The National Hospital for Neurology and Neurosurgery to perform these clinical and genetic studies on hereditary neuropathy. Cases were ascertained through either the genetic peripheral nerve clinic run by one of the authors (MMR), the neurogenetics clinic (HH) (NHNN), and those patients with characterized neuropathies referred for slit lamp microscopy and pupillography. In addition to neurological examination, all patients were assessed for autonomic features in their history and cardiac examination and lying and standing blood pressure calculations were carried out. One hundred and thirty one patients (57 male/74 female patients, aged 7–85 years) were recruited to the study. In the patients studied, the ocular history was reviewed to exclude patients with a blind eye, trauma, iritis, and conditions that could also affect pupil size and function. The medication history was also reviewed to exclude patients on narcotics, anticholinergics, pilocarpine eyedrops, and other drugs that could affect the pupils. The clinical diagnoses, sex distribution, and age range are given in Table 1.

Table 1 Clinical diagnoses and pupil abnormalities of patients studied
Full size table

Genetic diagnoses

DNA was extracted from blood samples obtained with informed consent from patients. DNA analysis was carried out in 126 patients; in 5 patients analysis could not be completed. The chromosome 17 duplication and deletion analysis and genetic sequencing of other genes has been described previously, but primers and methods are available on request23 (http://www.molgen.ua.ac.be/CMT).

Pupillography

Pupil diameters and their responses to light and accommodation (near) were recorded with a Whittaker/Applied Science Laboratories infrared television pupillometer as described previously.24 Bilateral recordings were made wherever possible and darkness anisocoria and right–left diameter differences recorded. For clarity and statistical analysis, the remaining measurements for only one eye are presented (right if available) per patient. Reduced pupil diameters in darkness and bilateral Horner's syndrome were taken as indicators of sympathetic dysfunction, whereas reduced light reflexes and/or mydriasis in the light, bilateral pupillotonia, and light–near dissociation as indicators of parasympathetic dysfunction.

Statistics

All measurements were compared with those obtained under identical conditions in 315 healthy subjects (172 male/143 female subjects) aged 16–82 years as described and reported previously.24

All patients were subjected to multiple pupil function tests, seven of which define quantitative abnormality as a value lying outside one 97.5% confidence interval. Within this scenario, in the assessment of individual patients, single quantitative abnormalities have a probability value of 0.150 and they have been ignored. Two abnormalities with a value P=0.012 and single abnormalities lying outside one 99.5% confidence interval (P=0.034) have been included. Bilateral tonic pupils and bilateral Horner's syndrome do not occur in healthy subjects; such abnormalities have been included.

Results

The clinical diagnosis and the pupil abnormalities in the 131 patients are summarized in Table 1. The genetic defects identified along with the individual patient findings are given in detail in Tables 2 and 3. Table 4 shows a comparison of pupil abnormalities in the different groups. Refsum's disease was diagnosed in 11 patients, based on raised phytanic acid. In the remaining 120 cases, 84 had a genetic diagnosis (70%). Overall, at least one significant pupil abnormality was found in 45/131 (34.4%) patients. We discuss only selected results as a great deal of information is given in the tables.

Table 2 Inherited neuropathy patients with abnormal pupils
Full size table
Table 3 Patients with inherited neuropathy and normal pupils
Full size table
Table 4 Pupil differences between inherited neuropathy diagnostic groups
Full size table

Familial amyloid polyneuropathy

In the FAP group, we have found a high prevalence of pupil abnormality in the 16 patients studied (Tables 1, 2 and 3). Two had bilateral tonic pupils with anisocoria, reduced light reflexes, but normal near responses. Of the other 14 patients, 8 had bilateral Horner's syndrome with redilatation lag (Figure 1). The remaining six had normal pupils. Out of the 10 patients with pupil abnormality, 8 had signs of autonomic dysfunction elsewhere, mainly blood pressure and cardiac abnormalities (Tables 2 and 3).

Figure 1
figure 1

Dark diameter (graphs on the left), the relationship between age and pupil dark diameter in healthy control subjects (top—normals), all patients with inherited neuropathy included in this study (middle—all patients), and patients with Refsum's disease (bottom—Refsum). Redilatation, relationship between light reflex amplitude, and redilatation time in healthy control subjects (top), all patients with inherited neuropathy (middle), and patients with familial amyloid neuropathy (bottom—FAP).

Full size image

Charcot Marie Tooth disease

In the CMT group, 12 of the 69 patients had abnormal pupils (Tables 2 and 3). Of the 12 pupil abnormalities, 8 were in the CMT2 group. The difference in occurrence of abnormality between clinical types is statistically significant (χ2=20.09, d.f.=7, P=0.005), with 42.1% of the CMT 2 patients having abnormal pupils. Three patients, all clinically CMT 2 (one in a recessive family), had bilateral Horner's syndrome. All were negative for Chromosome 17 duplication, two were negative for MPZ and Connexin mutations. In one patient, a mutation in the Mitofusin 2 gene was identified, case 13 (MFN2 Arg94Gln). This young female patient had a mild axonal neuropathy along with marked physiological anisocoria, outside the normal range (P<0.01), and the smaller pupil was abnormally small for her age. Unfortunately, no further MFN2 mutations have been identified to compare the pupil abnormalities. Fifty-seven patients had normal pupils. All six HNPP patients studied had normal pupils.

Refsum's disease

In Refsum's disease, 9 patients out of 11 had abnormally small pupils, 7 being below the 0.5% confidence limit (Figure 1). A total of 3/7 showed reduced light reflexes, 2/6 showed exaggerated near responses, and 3/6 showed light–near dissociation. None of the reflexes, although abnormal, was truly tonic. One patient had normal pupils. Tropicamide 0.5%, followed 30 min later by phenylephrine 10% eye drops, was instilled into one eye of 10 of these 11 patients. The final diameter reached was 6.00±0.64 mm (range 2.8–8.4 mm).

Hereditary sensory and autonomic neuropathy

The HSAN patients consisted of 13 (44.8%) out of 29 with abnormal pupils. In patients with the SPTLC1 C133W mutation, three out of four had abnormal pupils, and in the non-HSAN-I cases, there was a much higher incidence of pupil abnormalities. Both patients with HSAN-IV had abnormal pupils; both of these patients had a novel frameshift mutation in exon 15 of the TRKA gene (Table 2). In one patient with the TRKA mutation, pupillometry was impossible but photography revealed miosis below the 0.5% confidence interval of normal. The other had bilateral Horner's syndrome, also with severe miosis. The proband with HSAN-V (homozygous point mutation in exon 8 of the TRKA gene) had tonic pupils as did his father, who was a heterozygous carrier and was clinically normal otherwise.

Discussion

In this series of 131 patients with hereditary neuropathies, we provide an overview of the frequency and type of pupil defects. The neuropathy types were split into five groups, with the Refsum's disease and the autonomic neuropathies (FAP and HSAN) having the greatest frequency of defects. The severity of the autonomic neuropathy in the cases of FAP and those with HSAN was associated with a greater likelihood of having abnormal pupils, but this was not in all cases or all mutation types. This was most consistent in the patients with HSAN-III, -IV, and -V who all had pupil abnormalities and a severe neuropathy with autonomic features. There were cases of HSAN-I, HSAN-II, and FAP with moderate autonomic neuropathy but no pupil abnormalities. This is consistent with previous data on the small group of FAP and Refsum's patients but unreported in HSAN. In the parents available from recessive families with HSAN-II and HSAN-V, abnormal pupils were also recognized, suggesting that they were mild manifesting carriers.

The most frequent FAP group observed was the Ala60 TTR mutation (Irish type) seen in four cases. The main pupil abnormality seen was bilateral Horner's syndrome, reported also in light-chain-associated amyloidosis.25 The two cases with Met30 mutations did not have abnormal pupils, which was unexpected, as a previous study by Ando et al26 reported that 81% of Met30 patients had abnormal pupils early in their diagnosis.

The frequency of pupil abnormalities in CMT and HNPP was low and not associated with severity or mutation type, although one unexpected finding was the frequency of pupil abnormalities in CMT 2, especially as only one of the CMT 2 group with pupil abnormalities had an MPZ mutation. It is well recognized that MPZ mutations are a frequent cause of abnormal pupils in CMT 2 as in one of our cases.20, 23 It is, however, apparent that not all MPZ mutations cause pupil abnormalities, because we found normal pupils in cases with CMT1B and CMT2 and MPZ mutations. The majority of our CMT2 cases were also screened for the Mitofusin 2 gene, a frequent cause of CMT2A (Table 2). Only one case was found to have an MFN2 mutation and she had marked anisocoria. The phenotype spectrum of MFN2 mutations includes axonal neuropathy with optic atrophy, but there are no previous reports of pupil abnormalities.27 To confirm the association of abnormal pupils with MFN2 mutations, further cases need to be identified and examined. The frequency of pupil abnormalities in the CMT 2 group negative for MPZ and MFN2 suggests that the axonal neuropathy present in this form of CMT may in itself be associated with pupil abnormality or the unknown genetic defect(s) lead to abnormal deposits in the iris.

In the group with Refsum's disease, only one patient did not have a pupil abnormality. This case had an extremely strict diet and undetectable phytanic acid levels at the time of pupillography. The main abnormality found in the other patients was miosis.28, 29 The relative failure of the pupils of Refsum's disesase patients to dilate to phenylephrine suggests that in this condition, miosis is not due to an autonomic defect, but it is more likely due to a structural abnormality within the iris. This is consistent with previous pathological and electron microscopy examinations of the irides which have shown high concentrations of the phytanic acid lipid deposits in both sphincter and dilator muscles.30, 31 Any structural abnormality would correlate with the severity of the Refsum's disease and respond to diet restriction if started early enough. There are unfortunately no neuropathological studies on the irides of the other forms of inherited neuropathies in our cohort.

In summary, this is the first study of the pupil abnormalities found in a large group of inherited peripheral neuropathies and provides an overview of the frequency and type of defects seen but analysing this data we could not predict the genetic defect. In the FAP and HSAN (non-HSAN-I) groups, abnormal pupils were associated with a more severe neuropathy, but not in the other inherited neuropathy groups. This series along with the detailed tables will act as an important diagnostic aid in assessing these patients.